An intelligent performance diagnostic tool that connects to Tencent Cloud APM via the Model Context Protocol to analyze traces, spans, and system health.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install apm-performance-analysis
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install apm-performance-analysis using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is APM Performance Analysis?
The APM Performance Analysis skill leverages the Model Context Protocol (MCP) to bridge AI agents with remote APM servers. It enables developers to perform deep-dive diagnostics into business systems, providing real-time access to instance details, performance metrics, and call tracing. By utilizing SSE (Server-Sent Events) for connectivity, this skill allows for seamless interaction with application performance data without leaving the agent interface.
Designed for modern observability, it transforms raw performance data into actionable insights. Whether you are debugging high latency or inspecting flame graphs for resource bottlenecks, this Openclaw Skills integration provides a secure and structured way to maintain system health through an intelligent assistant.
APM Performance Analysis Use Cases
- Monitoring business system health and performance metrics through natural language queries.
- Tracing distributed call chains to identify latency bottlenecks in microservices.
- Analyzing flame graphs and span data to diagnose CPU or memory-intensive operations.
- Automated performance reporting and system status checks for DevOps teams.
- Securely accessing Tencent Cloud APM data using managed environment credentials.
How APM Performance Analysis Works
- The AI agent initializes a secure virtual environment to prevent global dependency conflicts.
- It checks for valid Tencent Cloud credentials (SecretId and SecretKey) within the local environment configuration.
- A connection is established to the remote APM MCP Server using the Server-Sent Events (SSE) protocol.
- The agent discovers available diagnostic tools by querying the MCP server's tool capability list.
- Upon user request, the agent constructs precise arguments and executes specific APM tools (e.g., fetching trace data or system metrics).
- The raw data returned via MCP is parsed directly by the AI and presented to the user through formatted Markdown tables and call trees.
APM Performance Analysis Setup
Follow these steps to configure the APM Performance Analysis skill for Openclaw Skills:
- Prepare Virtual Environment:
# Create and initialize the isolated Python environment
python scripts/venv_manager.py ensure
- Configure Credentials:
Create a
.envfile based onassets/.env.exampleand add your Tencent Cloud keys:
TENCENTCLOUD_SECRET_ID=your_id_here
TENCENTCLOUD_SECRET_KEY=your_key_here
APM_MCP_SSE_URL=https://mcp.tcop.woa.com/apm-console/ssen```nn3. **Test Connection:**
```bash
# List available performance tools
python scripts/venv_manager.py run scripts/mcp_client.py list-tools --output json
APM Performance Analysis Data Schema & Taxonomy
The skill manages performance data and configuration through a structured file hierarchy and secure logging:
| Component | Path / Detail | Description |
|---|---|---|
| Configuration | .env |
Stores sensitive API keys and MCP server URLs (chmod 600). |
| Logging | ./logs/apm_error.log |
Records error codes, RequestIds, and stack traces in JSON format. |
| Execution | scripts/ |
Contains the venv manager and the MCP SSE client. |
| Metadata | references/ |
Technical guides for interactions, credentials, and advanced MCP usage. |
| Data Format | Markdown / JSON | All tool outputs are rendered as Markdown tables or tree structures in-session. |
APM Performance Analysis Advanced Features
- Multi-platform mobile compatibility specifically designed for WorkBuddy and WeChat Enterprise, requiring no manual IDE confirmations.
- Automated flame graph rendering using hierarchical text or tree-shaped Markdown structures.
- Isolated execution environment via
venv_manager.pyto ensure local system stability. - Dynamic tool discovery allowing the agent to adapt to new APM capabilities added to the MCP server without code changes.
- Strict security protocols that prevent the creation of temporary scripts or dangerous file deletion commands.
SKILL.md
Loading
Related Openclaw Skills
Application O&M Report Generator
An automated diagnostic tool that converts Excel-based application failure logs into comprehensive, executive-ready Word O&M reports.
Blog Writer CN
A professional assistant for crafting high-quality, conversational Chinese technical blog posts that sound like a real person rather than an AI.
PAYMILL Integration Skill
A specialized integration for managing PAYMILL payment gateway data and automating financial workflows through a secure CLI-driven interface.
Sage Payroll
A powerful integration that automates Sage Payroll workflows, enabling seamless management of employees, payruns, and compliance data through a unified interface.
Sumo Smart Note
An automated note-taking skill that synchronizes development logs from your local workspace to a centralized shared notebook upon specific voice or text triggers.
GitHub
A powerful skill enabling AI agents to manage GitHub repositories, pull requests, and CI/CD pipelines using the official GitHub CLI.
Featured*
